Analysis of long-run benefits from international equity diversification between Taiwan and its major European trading partners: an empirical note

Abstract

This paper employs four cointegration test approaches, PO, HI, JJ and KSS, to test for pairwise long-run equilibrium relationships between Taiwan's stock price index and each of the stock price indexes of four European markets – French, German, Dutch, and British stock markets. The results from these four tests are robust and clearly consistent in suggesting that the Taiwan stock market is not pairwise cointegrated with the four European stock markets. This provides strong evidence that there exist long-run benefits for Taiwan investors diversifying in the equity markets of Taiwan's major European trading partners, France, Germany, Holland, and the UK, over the sample period considered from 6 January 1998 to 30 May 2002. These findings could be valuable to Taiwan individual investors and financial institutions holding long-run investment portfolios in the equity markets of France, Germany, Holland, and the UK.

Acknowledgements

The authors thank an anonymous referee and the editor Professor Mark Taylor for several helpful comments, suggestions, and time spent in reading this paper. These all make this paper more valuable and readable. Any errors that remain are our own.

Notes

¹ In order to catch the trading relationship between Taiwan and its major trading partners in Europe, Table 1 demonstrates the total trade volume, together with export and import volume for the period 2001 to 2003, to show their close trading relationship.

² PO, HI, JJ and KSS are abbreviations for the four cointegration tests: Phillips and Ouliaris (1990) multivariate trace statistic; Harris and Inder (1994) KPSS unit root-based; Johansen and Juselius (1990) maximum likelihood; and Kapetanios et al . (2003) nonlinear unit root-based approach, respectively.

³ Data is from the Securities Investment Trust and Consulting Association of ROC.

⁴ The null for KPSS is I(0), whereas it is I(1) for all other three tests, ADF, DF-GLS and PP.

⁵ According to Phillips and Ouliaris (1990, Table 4), the 5% critical value for the statistic for one explanatory variable is 55.2202.